How to correct/delete observations to create a Panel Data?

[Jose Williams]

Hello,

I am working with a panel data comprising of two waves written as Survey 1 and 2. The observations in survey 1 are correctly specified. However, in survey 2 some observations have been recorded twice or more due to some error.

As a result, I am getting more observations in the second wave.

IHDS1 |
(2005) or |
IHDS2 |
(2012) | Freq. Percent Cum.
------------+-----------------------------------
IHDS1 1 | 29,397 49.30 49.30
IHDS2 2 | 30,231 50.70 100.00
------------+-----------------------------------
Total | 59,628 100.00

I have attached an example of the data set using dataex. From dataex, id 8 is appearing four times instead of twice. Is there a way for me to correct this? I have tried 'duplicates report' but it did not work.

* Example generated by -dataex-. To install: ssc install dataex
clear
input int SURVEY double HHBASE float id1
1 1010201010  1
2 1010201010  1
1 1010201020  2
2 1010201020  2
1 1010201030  3
2 1010201030  3
1 1010201040  4
2 1010201040  4
1 1010201050  5
2 1010201050  5
1 1010201070  6
2 1010201070  6
1 1010201080  7
2 1010201080  7
1 1010201090  8
2 1010201090  8
2 1010201090  8
2 1010201090  8
1 1010201100  9
2 1010201100  9
1 1010201120 10
2 1010201120 10
1 1010201130 11
2 1010201130 11
1 1010201140 12
2 1010201140 12
1 1010201160 13
2 1010201160 13
1 1010201170 14
2 1010201170 14
1 1010201180 15
2 1010201180 15
1 1010201190 16
2 1010201190 16
1 1010201200 17
2 1010201200 17
1 1010202010 18
2 1010202010 18
1 1010202020 19
2 1010202020 19
1 1010202030 20
2 1010202030 20
1 1010202040 21
2 1010202040 21
1 1010202060 22
2 1010202060 22
1 1010202070 23
2 1010202070 23
1 1010202100 24
2 1010202100 24
1 1010202110 25
2 1010202110 25
1 1010202140 26
2 1010202140 26
1 1010202150 27
2 1010202150 27
1 1010202160 28
2 1010202160 28
1 1010202170 29
2 1010202170 29
1 1010202180 30
2 1010202180 30
1 1010202190 31
2 1010202190 31
1 1010202200 32
2 1010202200 32
1 1010203020 33
2 1010203020 33
1 1010203050 34
2 1010203050 34
1 1010203060 35
2 1010203060 35
1 1010203070 36
2 1010203070 36
1 1010203090 37
2 1010203090 37
1 1010203100 38
2 1010203100 38
2 1010203100 38
2 1010203100 38
1 1010203110 39
2 1010203110 39
1 1010203120 40
2 1010203120 40
1 1010203130 41
2 1010203130 41
1 1010203140 42
2 1010203140 42
1 1010203150 43
2 1010203150 43
1 1010203160 44
2 1010203160 44
1 1010203170 45
2 1010203170 45
1 1010203190 46
2 1010203190 46
1 1010204010 47
2 1010204010 47
1 1010204040 48
2 1010204040 48
end
label values SURVEY SURVEY
label def SURVEY 1 "IHDS1 1", modify
label def SURVEY 2 "IHDS2 2", modify

[Nick Cox]

See 12.1 in https://www.statalist.org/forums/help#stata for advice never to say just "did not work". What happened (or did not happen)? Why was it not what you wanted or expected?

duplicates report

works fine here, as does

duplicates drop

to clean up your data example.

Detail: "a panel dataset" (or data set) works fine as idiomatic English; "a panel data" not so well in my hearing.